pygpt-net 2.6.30__py3-none-any.whl → 2.6.32__py3-none-any.whl

pygpt_net/data/css/web-blocks.css

@@ -404,4 +404,10 @@ think {{
 }}
 .tips.visible {{
  opacity: 1;
+}}
+
+.extra-src-video-box,
+.video-player {{
+  max-width: 400px;
+  text-align: left;
 }}

pygpt_net/data/css/web-chatgpt.css

@@ -434,4 +434,10 @@ think {{
 
 .name-user {{
   display: none;
+}}
+
+.extra-src-video-box,
+.video-player {{
+  max-width: 400px;
+  text-align: left;
 }}

pygpt_net/data/css/web-chatgpt_wide.css

@@ -430,4 +430,10 @@ think {{
 
 .name-user {{
   display: none;
+}}
+
+.extra-src-video-box,
+.video-player {{
+  max-width: 400px;
+  text-align: left;
 }}

pygpt_net/data/locale/locale.de.ini

@@ -191,6 +191,7 @@ attachments_uploaded.tab = Hochgeladene Dateien
 audio.cache.clear.confirm = Sind Sie sicher, dass Sie alle zwischengespeicherten Audiodateien löschen möchten?
 audio.cache.clear.success = OK. Alle Audio-Cache-Dateien wurden gelöscht.
 audio.control.btn = Sprachsteuerung
+audio.loop = Schleife
 audio.magic_word.detected = Zauberwort erkannt!
 audio.magic_word.invalid = Kein Zauberwort :(
 audio.magic_word.please = Bitte das Zauberwort...
@@ -845,7 +846,7 @@ mode.agent_openai.tooltip = Fortgeschrittene Agenten (OpenAI)
 mode.agent.tooltip = Einfache Agenten (legacy)
 mode.assistant = Assistent
 mode.assistant.tooltip = Chat mittels Assistants API
-mode.audio = Chat mit Audio
+mode.audio = Realtime + audio
 mode.chat = Chat
 mode.chat.tooltip = Chatmodus (Standard)
 mode.completion = Vervollständigung
@@ -854,7 +855,7 @@ mode.computer = Computerbenutzung
 mode.computer.tooltip = Computerbenutzung (Maus, Tastatur, Navigation)
 mode.expert = Experten (Kooperation)
 mode.expert.tooltip = Experten zum Hintergrund rufen
-mode.img = Bild
+mode.img = Bild und Video
 mode.img.tooltip = Bildgenerierung mit DALL-E
 mode.langchain = Langchain
 mode.langchain.tooltip = Chat mit Modellen von Langchain
@@ -1127,6 +1128,8 @@ settings.audio.input.stop_interval.desc = Intervall in Sekunden für automatisch
 settings.audio.input.timeout = Aufnahme-Zeitüberschreitung
 settings.audio.input.timeout.continuous = Zeitüberschreitung im Dauerbetrieb aktivieren
 settings.audio.input.timeout.desc = Zeitüberschreitung (Sekunden) für automatische Stop-Aufzeichnung, 0 zum Deaktivieren, Standard: 120
+settings.audio.input.vad.prefix = VAD-Präfix-Puffer (in ms)
+settings.audio.input.vad.silence = VAD-Endstille (in ms)
 settings.audio.output.backend = Backend für Audioausgabe
 settings.audio.output.backend.desc = Wählen Sie das Backend für die Audioausgabe.
 settings.audio.output.device = Gerät für Audioausgabe
@@ -1197,9 +1200,10 @@ settings.frequency_penalty = Frequenzstrafe
 settings.func_call.native = Native API-Funktionsaufrufe verwenden
 settings.func_call.native.desc = Wenn aktiviert, benutzt die Anwendung native API-Funktionsaufrufe anstelle des internen pygpt-Formats und der unten stehenden Befehlsprompten. Nur Chat- und Assistenz-Modi.
 settings.img_dialog_open = Bild-Dialog nach Generierung öffnen (Bildmodus)
-settings.img_prompt_model = DALL-E: Modell zur Prompterzeugung
-settings.img_quality = DALL-E: image quality
-settings.img_resolution = DALL-E: Bildgröße
+settings.img_prompt_model = Modell zur Prompterzeugung
+settings.img_prompt_model.desc = LLM zur Verfeinerung Ihrer Eingabeaufforderung vor der Bilderzeugung (nicht das Bildmodell)
+settings.img_quality = image quality
+settings.img_resolution = Bildgröße
 settings.layout.animation.disable = Animationen deaktivieren
 settings.layout.animation.disable.desc = Deaktiviert Layout-Animationen, wie animierte Ladegeräte usw.
 settings.layout.density = Layoutdichte
@@ -1294,8 +1298,10 @@ settings.prompt.ctx.auto_summary.user = Kontext: Auto-Zusammenfassung (Benutzern
 settings.prompt.ctx.auto_summary.user.desc = Platzhalter: {input}, {output}
 settings.prompt.expert = Experte: Masteraufforderung
 settings.prompt.expert.desc = Anweisung (Systemaufforderung) für den Master-Experten, wie man Sklavenexperten handhabt. Anweisungen für Sklavenexperten werden aus ihren Voreinstellungen gegeben.
-settings.prompt.img = DALL-E: Bildgenerierung
+settings.prompt.img = Bildgenerierung
 settings.prompt.img.desc = Aufforderung zur Erzeugung von Anweisungen für DALL-E (falls Rohmodus deaktiviert ist). Nur im Bildmodus.
+settings.prompt.video = Videogenerierung
+settings.prompt.video.desc = Eingabeaufforderung zur Erzeugung von Eingabeaufforderungen für das Videomodell (wenn der Rohmodus deaktiviert ist). Nur Bild-/Videomodus.
 settings.remote_tools.code_interpreter = Code-Interpreter
 settings.remote_tools.code_interpreter.desc = Aktivieren Sie das `code_interpreter` Remote-Tool im Chat-Modus / über OpenAI Responses API.
 settings.remote_tools.file_search = File search
@@ -1339,7 +1345,9 @@ settings.section.ctx = Kontext
 settings.section.debug = Fehlerbehebung
 settings.section.files = Dateien und Anhänge
 settings.section.general = Allgemein
-settings.section.images = Bilder
+settings.section.images = Bilder und Videos
+settings.section.images.image = Bild
+settings.section.images.video = Video
 settings.section.layout = Layout
 settings.section.llama_index = Indizes / LlamaIndex
 settings.section.llama-index.chat = Chat
@@ -1367,6 +1375,22 @@ settings.upload.data_dir.desc = Włącz, aby przechowywać wszystko w jednym kat
 settings.upload.store = Speichere Anhänge im Arbeitsverzeichnis im Upload-Ordner
 settings.upload.store.desc = Aktiviere das Speichern einer lokalen Kopie hochgeladener Anhänge zur späteren Verwendung
 settings.use_context = Kontext (Erinnerung) verwenden
+settings.video.aspect_ratio = Seitenverhältnis
+settings.video.aspect_ratio.desc = Seitenverhältnis des Rahmens (z.B. 16:9, 9:16, 1:1); Verfügbarkeit hängt vom ausgewählten Modell ab
+settings.video.duration = Videodauer
+settings.video.duration.desc = Clip-Länge in Sekunden; Grenzen können je nach Modell variieren
+settings.video.fps = FPS
+settings.video.fps.desc = Bilder pro Sekunde (z.B. 24, 25, 30); kann vom Modell gerundet oder ignoriert werden
+settings.video.generate_audio = Audio generieren
+settings.video.generate_audio.desc = Bei Unterstützung durch das Modell synthetisiertes Hintergrundaudio einschließen
+settings.video.negative_prompt = Negative Eingabeaufforderung
+settings.video.negative_prompt.desc = Wörter oder Sätze, die im Ergebnis vermieden werden sollen (durch Kommas getrennt)
+settings.video.prompt_model = Modell zur Verbesserung von Eingabeaufforderungen
+settings.video.prompt_model.desc = LLM zur Verfeinerung Ihrer Eingabeaufforderung vor der Videogenerierung (nicht das Videomodell)
+settings.video.resolution = Videoauflösung
+settings.video.resolution.desc = Zielausgabeauflösung (z.B. 720p, 1080p); Verfügbarkeit hängt vom Modell ab
+settings.video.seed = Startwert
+settings.video.seed.desc = Optionaler Zufallsstartwert für reproduzierbare Ergebnisse; leer lassen für zufällig
 settings.vision.capture.auto = Automatisches Erfassen
 settings.vision.capture.enabled = Kamera
 settings.vision.capture.height = Kamerabildhöhe (px)
@@ -1526,6 +1550,10 @@ updater.check.launch = Beim Start prüfen
 update.released = Build
 update.snap = Zum Snap Store gehen
 update.title = Nach Updates suchen
+vid.status.downloading = Video wird heruntergeladen... bitte warten...
+vid.status.generating = Video wird generiert von
+vid.status.prompt.error = Fehler bei der Verbesserungsaufforderung
+vid.status.prompt.wait = Vorbereitung der Eingabeaufforderung... bitte warten...
 vision.capture.auto = Automatische Erfassung
 vision.capture.auto.click = Automatisches Erfassen ist aktiviert!
 vision.capture.auto.label = Automatische Erfassung aktiviert

pygpt_net/data/locale/locale.en.ini

@@ -188,9 +188,11 @@ attachments_uploaded.clear.confirm = WARNING: are you sure you want to delete al
 attachments_uploaded.delete.confirm = WARNING: are you sure you want to delete this file from the remote server?
 attachments_uploaded.sync.tip = Tip: click on 'Sync' to retrieve the file list from OpenAI
 attachments_uploaded.tab = Uploaded
+audio.auto_turn = Auto (VAD)
 audio.cache.clear.confirm = Are you sure you want to delete all cached audio files?
 audio.cache.clear.success = OK. All audio cache files cleared.
 audio.control.btn = Voice control
+audio.loop = Loop
 audio.magic_word.detected = Magic word detected!
 audio.magic_word.invalid = Not a magic word :(
 audio.magic_word.please = Magic word please...
@@ -845,7 +847,7 @@ mode.agent_openai.tooltip = Advanced agents (OpenAI)
 mode.agent.tooltip = Simple agents (legacy)
 mode.assistant = Assistants
 mode.assistant.tooltip = Chat using Assistants API
-mode.audio = Chat with Audio
+mode.audio = Realtime + audio
 mode.chat = Chat
 mode.chat.tooltip = Chat mode (default)
 mode.completion = Completion
@@ -854,8 +856,8 @@ mode.computer = Computer use
 mode.computer.tooltip = Computer use (mouse, keyboard, navigation)
 mode.expert = Experts (Co-op)
 mode.expert.tooltip = Experts to call in the background
-mode.img = Image
-mode.img.tooltip = Image generation using DALL-E
+mode.img = Image and video
+mode.img.tooltip = Image generation
 mode.langchain = Langchain
 mode.langchain.tooltip = Chat with models provided by Langchain
 model.ctx = Context tokens
@@ -1119,7 +1121,15 @@ settings.api_key.voyage.desc = Required for the Voyage API - embeddings for Anth
 settings.api_key.xai = xAI API KEY
 settings.api_key.xai.desc = Required for the xAI API and Grok models.
 settings.api_native_google = Use native API SDK
+settings.api_native_google.app_credentials =  Google Application credentials (path)
+settings.api_native_google.app_credentials.desc = Absolute path to credentials.json, e.g. /home/user/credentials.json
+settings.api_native_google.cloud_location =  Google Cloud location
+settings.api_native_google.cloud_location.desc = Provide your Google Cloud project location, default: us-central1
+settings.api_native_google.cloud_project = Google Cloud project
+settings.api_native_google.cloud_project.desc = Provide your Google Cloud project name
 settings.api_native_google.desc = Use native GenAI SDK instead of compatible OpenAI client
+settings.api_native_google.use_vertex = Use VertexAI
+settings.api_native_google.use_vertex.desc = Enable to use VertexAI in Google GenAI SDK
 settings.api_proxy = Proxy address
 settings.api_proxy.desc = Optional, proxy for OpenAI API, e.g. http://proxy.example.com or socks5://user:pass@host:port
 settings.api_use_responses = Use Responses API in Chat mode
@@ -1147,6 +1157,8 @@ settings.audio.input.stop_interval.desc = Interval in seconds for auto-transcrib
 settings.audio.input.timeout = Recording timeout
 settings.audio.input.timeout.continuous = Enable timeout in continuous mode
 settings.audio.input.timeout.desc = Timeout (seconds) for auto-stop recording, 0 to disable, default: 120
+settings.audio.input.vad.prefix = VAD prefix padding (in ms)
+settings.audio.input.vad.silence = VAD end silence (in ms)
 settings.audio.output.backend = Audio Output Backend
 settings.audio.output.backend.desc = Select the audio output backend.
 settings.audio.output.device = Audio Output Device
@@ -1219,9 +1231,10 @@ settings.frequency_penalty = Frequency Penalty
 settings.func_call.native = Use native API function calls
 settings.func_call.native.desc = If enabled, the application will use native API function calls instead of the internal pygpt format and the command prompts from below will not be used. Chat and Assistants modes ONLY.
 settings.img_dialog_open = Open image dialog after generation (Image mode)
-settings.img_prompt_model = DALL-E: prompt generation model
-settings.img_quality = DALL-E: image quality
-settings.img_resolution = DALL-E: image size
+settings.img_prompt_model = Prompt generation model
+settings.img_prompt_model.desc = LLM used to refine your prompt before image generation (not the image model)
+settings.img_quality = Image quality
+settings.img_resolution = Image size
 settings.layout.animation.disable = Disable animations
 settings.layout.animation.disable.desc = Disables layout animations, like animated loaders, etc.
 settings.layout.density = Layout density
@@ -1318,26 +1331,30 @@ settings.prompt.ctx.auto_summary.user = Context: auto-summary (user message)
 settings.prompt.ctx.auto_summary.user.desc = Placeholders: {input}, {output}
 settings.prompt.expert = Expert: Master prompt
 settings.prompt.expert.desc = Instruction (system prompt) for Master expert on how to handle slave experts. Instructions for slave experts are given from their presets.
-settings.prompt.img = DALL-E: image generation
-settings.prompt.img.desc = Prompt for generating prompts for DALL-E (if raw-mode is disabled). Image mode only.
+settings.prompt.img = Image generation
+settings.prompt.img.desc = Prompt for generating prompts for image model (if raw-mode is disabled). Image / Video modes only.
+settings.prompt.video = Video generation
+settings.prompt.video.desc = Prompt for generating prompts for video model (if raw-mode is disabled). Image / Videos mode only.
 settings.remote_tools.code_interpreter = Code Interpreter
-settings.remote_tools.code_interpreter.desc = Enable `code_interpreter` remote tool in Chat mode / via OpenAI Responses API.
+settings.remote_tools.code_interpreter.desc = Enable `code_interpreter` remote tool - Responses API only.
 settings.remote_tools.file_search = File search
 settings.remote_tools.file_search.args = File search vector store IDs
 settings.remote_tools.file_search.args.desc = Vector store IDs, separated by comma (,)
-settings.remote_tools.file_search.desc = Enable `file_search` remote tool in Chat mode / via OpenAI Responses API.
+settings.remote_tools.file_search.desc = Enable `file_search` remote tool - Responses API only.
 settings.remote_tools.google.code_interpreter = Code Interpreter
-settings.remote_tools.google.code_interpreter.desc = Enable Code Interpreter remote tool in Chat mode.
-settings.remote_tools.google.web_search = Google Web Search
-settings.remote_tools.google.web_search.desc = Enable Google Search remote tool in Chat mode.
+settings.remote_tools.google.code_interpreter.desc = Enable Code Interpreter remote tool.
+settings.remote_tools.google.url_ctx = URL Context
+settings.remote_tools.google.url_ctx.desc = Enable URL Context remote tool.
+settings.remote_tools.google.web_search = Google Search
+settings.remote_tools.google.web_search.desc = Enable Google Search remote tool.
 settings.remote_tools.image = Image generation
-settings.remote_tools.image.desc = Enable `image_generation` remote tool in Chat mode / via OpenAI Responses API.
+settings.remote_tools.image.desc = Enable `image_generation` remote tool - Responses API only.
 settings.remote_tools.mcp = Remote MCP
 settings.remote_tools.mcp.args = Remote MCP configuration
 settings.remote_tools.mcp.args.desc = Configuration in JSON format (will be used in request)
-settings.remote_tools.mcp.desc = Enable `mcp` remote tool in Chat mode / via OpenAI Responses API.
+settings.remote_tools.mcp.desc = Enable `mcp` remote tool - Responses API only.
 settings.remote_tools.web_search = Web Search
-settings.remote_tools.web_search.desc = Enable `web_search` remote tool in Chat mode / via OpenAI Responses API.
+settings.remote_tools.web_search.desc = Enable `web_search` remote tool - Responses API only.
 settings.render.code_syntax = Code syntax highlight
 settings.render.engine = Rendering engine
 settings.render.open_gl = OpenGL hardware acceleration
@@ -1369,7 +1386,9 @@ settings.section.ctx = Context
 settings.section.debug = Debug
 settings.section.files = Files and attachments
 settings.section.general = General
-settings.section.images = Images
+settings.section.images = Images and video
+settings.section.images.image = Image
+settings.section.images.video = Video
 settings.section.layout = Layout
 settings.section.llama_index = Indexes / LlamaIndex
 settings.section.llama-index.chat = Chat
@@ -1398,6 +1417,22 @@ settings.upload.data_dir.desc = Enable to store everything in a single data dire
 settings.upload.store = Store attachments in the workdir upload directory
 settings.upload.store.desc = Enable to store a local copy of uploaded attachments for future use
 settings.use_context = Use context (memory)
+settings.video.aspect_ratio = Aspect ratio
+settings.video.aspect_ratio.desc = Frame aspect ratio (e.g., 16:9, 9:16, 1:1); availability depends on the selected model
+settings.video.duration = Video duration
+settings.video.duration.desc = Clip length in seconds; limits may vary by model
+settings.video.fps = FPS
+settings.video.fps.desc = Frames per second (e.g., 24, 25, 30); may be rounded or ignored by the model
+settings.video.generate_audio = Generate audio
+settings.video.generate_audio.desc = Include synthesized background audio if supported by the model
+settings.video.negative_prompt = Negative prompt
+settings.video.negative_prompt.desc = Words or phrases to avoid in the output (comma-separated)
+settings.video.prompt_model = Prompt enhancement model
+settings.video.prompt_model.desc = LLM used to refine your prompt before video generation (not the video model)
+settings.video.resolution = Video resolution
+settings.video.resolution.desc = Target output resolution (e.g., 720p, 1080p); availability depends on the model
+settings.video.seed = Seed
+settings.video.seed.desc = Optional random seed for reproducible results; leave empty for random
 settings.vision.capture.auto = Auto capture
 settings.vision.capture.enabled = Camera
 settings.vision.capture.height = Camera height (px)
@@ -1559,6 +1594,10 @@ updater.check.launch = Check on launch
 update.released = build
 update.snap = Go to Snap Store
 update.title = Check for Updates
+vid.status.downloading = Downloading video... please wait...
+vid.status.generating = Generating video from
+vid.status.prompt.error = Enhancement prompt error occured
+vid.status.prompt.wait = Preparing prompt... please wait...
 vision.capture.auto = Auto capture
 vision.capture.auto.click = Auto-capture is enabled!
 vision.capture.auto.label = Auto capture enabled

pygpt_net/data/locale/locale.es.ini

@@ -191,6 +191,7 @@ attachments_uploaded.tab = Archivos subidos
 audio.cache.clear.confirm = ¿Está seguro de querer eliminar todos los archivos de audio en caché?
 audio.cache.clear.success = OK. Todos los archivos de caché de audio fueron eliminados.
 audio.control.btn = Control de voz
+audio.loop = Bucle
 audio.magic_word.detected = ¡Palabra mágica detectada!
 audio.magic_word.invalid = No es una palabra mágica :(
 audio.magic_word.please = Por favor, una palabra mágica...
@@ -846,7 +847,7 @@ mode.agent_openai.tooltip = Agentes avanzados (OpenAI)
 mode.agent.tooltip = Agentes simples (legacy)
 mode.assistant = Asistente
 mode.assistant.tooltip = Chatear usando la API de Asistentes
-mode.audio = Chat con audio
+mode.audio = Realtime + audio
 mode.chat = Chat
 mode.chat.tooltip = Modo de chat (predeterminado)
 mode.completion = Finalización
@@ -855,7 +856,7 @@ mode.computer = Uso de la computadora
 mode.computer.tooltip = Uso de la computadora (ratón, teclado, navegación)
 mode.expert = Expertos (cooperación)
 mode.expert.tooltip = Llamar a expertos en segundo plano
-mode.img = Imagen
+mode.img = Imagen y video
 mode.img.tooltip = Generar imagen usando DALL-E
 mode.langchain = Langchain
 mode.langchain.tooltip = Chatear con modelos proporcionados por Langchain
@@ -1128,6 +1129,8 @@ settings.audio.input.stop_interval.desc = Intervalo en segundos para transcribir
 settings.audio.input.timeout = Tiempo de espera de grabación
 settings.audio.input.timeout.continuous = Habilitar tiempo de espera en modo continuo
 settings.audio.input.timeout.desc = Tiempo de espera (segundos) para detener la grabación automáticamente, 0 para deshabilitar, por defecto: 120
+settings.audio.input.vad.prefix = Relleno de prefijo VAD (en ms)
+settings.audio.input.vad.silence = Silencio de final VAD (en ms)
 settings.audio.output.backend = Backend para la salida de audio
 settings.audio.output.backend.desc = Selecciona el backend para la salida de audio.
 settings.audio.output.device = Dispositivo para la salida de audio
@@ -1198,9 +1201,10 @@ settings.frequency_penalty = Penalización de frecuencia
 settings.func_call.native = Usar llamadas a funciones API nativas
 settings.func_call.native.desc = Si está habilitado, la aplicación usará llamadas a funciones API nativas en lugar del formato pygpt interno y los prompt de comandos a continuación no se usarán. Solo modos de Chat y Asistentes.
 settings.img_dialog_open = Abrir diálogo de imagen después de generar (Modo imagen)
-settings.img_prompt_model = DALL-E: modelo de generación de indicaciones
-settings.img_quality = DALL-E: calidad de imagen
-settings.img_resolution = DALL-E: tamaño de imagen
+settings.img_prompt_model = Modelo de generación de indicaciones
+settings.img_prompt_model.desc = LLM utilizado para refinar su indicación antes de la generación de imágenes (no el modelo de imágenes)
+settings.img_quality = Calidad de imagen
+settings.img_resolution = Tamaño de imagen
 settings.layout.animation.disable = Desactivar animaciones
 settings.layout.animation.disable.desc = Desactivar animaciones de diseño, como cargadores animados, etc.
 settings.layout.density = Densidad de la disposición
@@ -1295,8 +1299,10 @@ settings.prompt.ctx.auto_summary.user = Contexto: resumen automático (mensaje d
 settings.prompt.ctx.auto_summary.user.desc = Marcadores de posición: {input}, {output}
 settings.prompt.expert = Experto: Master prompt
 settings.prompt.expert.desc = Instrucción (prompt del sistema) para el experto Master cómo manejar a los expertos subordinados. Las instrucciones para los expertos subordinados se dan desde sus presets.
-settings.prompt.img = DALL-E: generación de imagen
+settings.prompt.img = Generación de imagen
 settings.prompt.img.desc = Mensaje para generar comandos para DALL-E (si el modo crudo está desactivado). Solo modo de imagen.
+settings.prompt.video = Generación de video
+settings.prompt.video.desc = Indicaciones para generar indicaciones para el modelo de video (si el modo sin formato está desactivado). Solo modo Imagen/Videos.
 settings.remote_tools.code_interpreter = Intérprete de Código
 settings.remote_tools.code_interpreter.desc = Habilitar herramienta remota `code_interpreter` en modo Chat / vía API de Respuestas de OpenAI.
 settings.remote_tools.file_search = File search
@@ -1340,7 +1346,9 @@ settings.section.ctx = Contexto
 settings.section.debug = Depuración
 settings.section.files = Archivos y adjuntos
 settings.section.general = General
-settings.section.images = Imágenes
+settings.section.images = Imágenes y vídeo
+settings.section.images.image = Imagen
+settings.section.images.video = Video
 settings.section.layout = Diseño
 settings.section.llama_index = Índices / LlamaIndex
 settings.section.llama-index.chat = Chat
@@ -1368,6 +1376,22 @@ settings.upload.data_dir.desc = Habilite para almacenar todo en un único direct
 settings.upload.store = Almacenar adjuntos en el directorio de subida del directorio de trabajo
 settings.upload.store.desc = Habilitar el almacenamiento de una copia local de los adjuntos subidos para su uso futuro
 settings.use_context = Utilizar contexto (memoria)
+settings.video.aspect_ratio = Relación de aspecto
+settings.video.aspect_ratio.desc = Relación de aspecto del cuadro (por ejemplo, 16:9, 9:16, 1:1); la disponibilidad depende del modelo seleccionado
+settings.video.duration = Duración de video
+settings.video.duration.desc = Longitud del clip en segundos; los límites pueden variar según el modelo
+settings.video.fps = FPS
+settings.video.fps.desc = Fotogramas por segundo (por ejemplo, 24, 25, 30); puede ser redondeado o ignorado por el modelo
+settings.video.generate_audio = Generar audio
+settings.video.generate_audio.desc = Incluir audio de fondo sintetizado si es compatible con el modelo
+settings.video.negative_prompt = Indicaciones negativas
+settings.video.negative_prompt.desc = Palabras o frases a evitar en el resultado (separadas por comas)
+settings.video.prompt_model = Modelo de mejora de indicación
+settings.video.prompt_model.desc = LLM utilizado para refinar su indicación antes de la generación de video (no el modelo de video)
+settings.video.resolution = Resolución de video
+settings.video.resolution.desc = Resolución de salida objetivo (por ejemplo, 720p, 1080p); la disponibilidad depende del modelo
+settings.video.seed = Semilla
+settings.video.seed.desc = Semilla aleatoria opcional para resultados reproducibles; déjelo vacío para aleatorio
 settings.vision.capture.auto = Captura automática
 settings.vision.capture.enabled = Cámara
 settings.vision.capture.height = Altura de captura de cámara (px)
@@ -1527,6 +1551,10 @@ updater.check.launch = Comprobar al iniciar
 update.released = Versión
 update.snap = Ir a Snap Store
 update.title = Buscando actualizaciones
+vid.status.downloading = Descargando video... por favor espere...
+vid.status.generating = Generando video desde
+vid.status.prompt.error = Ocurrió un error en la indicación de mejora
+vid.status.prompt.wait = Preparando indicación... por favor espere...
 vision.capture.auto = Captura automática
 vision.capture.auto.click = ¡La captura automática está habilitada!
 vision.capture.auto.label = Captura automática habilitada

pygpt_net/data/locale/locale.fr.ini

@@ -191,6 +191,7 @@ attachments_uploaded.tab = Fichiers téléchargés
 audio.cache.clear.confirm = Êtes-vous sûr de vouloir supprimer tous les fichiers audio en cache ?
 audio.cache.clear.success = OK. Tous les fichiers de cache audio ont été effacés.
 audio.control.btn = Contrôle vocal
+audio.loop = Boucle
 audio.magic_word.detected = Mot magique détecté !
 audio.magic_word.invalid = Pas un mot magique :(
 audio.magic_word.please = S'il vous plaît, le mot magique...
@@ -845,7 +846,7 @@ mode.agent_openai.tooltip = Agents avancés (OpenAI)
 mode.agent.tooltip = Agents simples (legacy)
 mode.assistant = Assistant
 mode.assistant.tooltip = Discuter via l'API des Assistants
-mode.audio = Chat avec audio
+mode.audio = Realtime + audio
 mode.chat = Chat
 mode.chat.tooltip = Mode chat (par défaut)
 mode.completion = Complétion
@@ -854,7 +855,7 @@ mode.computer = Utilisation de l'ordinateur
 mode.computer.tooltip = Utilisation de l'ordinateur (souris, clavier, navigation)
 mode.expert = Experts (co-opération)
 mode.expert.tooltip = Appeler des experts en arrière-plan
-mode.img = Image
+mode.img = Image et vidéo
 mode.img.tooltip = Génération d'image avec DALL-E
 mode.langchain = Langchain
 mode.langchain.tooltip = Discussion avec les modèles fournis par Langchain
@@ -1127,6 +1128,8 @@ settings.audio.input.stop_interval.desc = Intervalle en secondes pour la transcr
 settings.audio.input.timeout = Délai d'enregistrement
 settings.audio.input.timeout.continuous = Activer le délai en mode continu
 settings.audio.input.timeout.desc = Délai (secondes) pour l'arrêt automatique de l'enregistrement, 0 pour désactiver, par défaut : 120
+settings.audio.input.vad.prefix = Padding du préfixe VAD (en ms)
+settings.audio.input.vad.silence = Silence de fin VAD (en ms)
 settings.audio.output.backend = Backend pour la sortie audio
 settings.audio.output.backend.desc = Sélectionnez le backend pour la sortie audio.
 settings.audio.output.device = Périphérique pour la sortie audio
@@ -1197,9 +1200,10 @@ settings.frequency_penalty = Pénalité de fréquence
 settings.func_call.native = Utiliser les appels de fonction API natives
 settings.func_call.native.desc = Si activé, l'application utilisera les appels de fonction API natives au lieu du format interne de pygpt et les commandes prompt ci-dessous ne seront pas utilisées. Modes uniquement Chat et Assistants.
 settings.img_dialog_open = Ouvrir la boîte de dialogue d'image après la génération (Mode image)
-settings.img_prompt_model = DALL-E : modèle de génération d'invite
-settings.img_quality = DALL-E: qualité d'image
-settings.img_resolution = DALL-E : taille de l'image
+settings.img_prompt_model = Modèle de génération d'invite
+settings.img_prompt_model.desc = LLM utilisé pour affiner votre invite avant la génération d'images (pas le modèle d'image)
+settings.img_quality = Qualité d'image
+settings.img_resolution = Taille de l'image
 settings.layout.animation.disable = Désactiver les animations
 settings.layout.animation.disable.desc = Désactive les animations de mise en page, comme les chargeurs animés, etc.
 settings.layout.density = Densité de la disposition
@@ -1294,8 +1298,10 @@ settings.prompt.ctx.auto_summary.user = Contexte: résumé automatique (message
 settings.prompt.ctx.auto_summary.user.desc = Espaces réservés: {input}, {output}
 settings.prompt.expert = Expert : Master prompt
 settings.prompt.expert.desc = Instruction (prompt système) pour l'expert Master sur comment gérer les experts esclaves. Les instructions pour les experts esclaves sont données à partir de leurs presets.
-settings.prompt.img = DALL-E: génération d'image
+settings.prompt.img = Génération d'image
 settings.prompt.img.desc = Prompt pour générer des commandes pour DALL-E (si le mode brut est désactivé). Mode image uniquement.
+settings.prompt.video = Génération de vidéo
+settings.prompt.video.desc = Invite pour générer des invites pour le modèle vidéo (si le mode brut est désactivé). Mode Image/Vidéo uniquement.
 settings.remote_tools.code_interpreter = Interpréteur de code
 settings.remote_tools.code_interpreter.desc = Activer l'outil distant `code_interpreter` en mode Chat/ via OpenAI Responses API.
 settings.remote_tools.file_search = File search
@@ -1339,7 +1345,9 @@ settings.section.ctx = Contexte
 settings.section.debug = Débogage
 settings.section.files = Fichiers et pièces jointes
 settings.section.general = Général
-settings.section.images = Images
+settings.section.images = Images et vidéos
+settings.section.images.image = Image
+settings.section.images.video = Vidéo
 settings.section.layout = Mise en page
 settings.section.llama_index = Indexes / LlamaIndex
 settings.section.llama-index.chat = Chat
@@ -1367,6 +1375,22 @@ settings.upload.data_dir.desc = Activez pour stocker tout dans un seul répertoi
 settings.upload.store = Stocker les pièces jointes dans le répertoire d'upload du répertoire de travail
 settings.upload.store.desc = Activer le stockage d'une copie locale des pièces jointes téléchargées pour une utilisation future
 settings.use_context = Utiliser le contexte (mémoire)
+settings.video.aspect_ratio = Ratio d'aspect
+settings.video.aspect_ratio.desc = Ratio d'aspect du cadre (ex. : 16:9, 9:16, 1:1); disponibilité dépend du modèle sélectionné
+settings.video.duration = Durée de la vidéo
+settings.video.duration.desc = Longueur du clip en secondes; les limites peuvent varier selon le modèle
+settings.video.fps = FPS
+settings.video.fps.desc = Images par seconde (ex. : 24, 25, 30); peut être arrondi ou ignoré par le modèle
+settings.video.generate_audio = Générer de l'audio
+settings.video.generate_audio.desc = Inclure un fond audio synthétisé si pris en charge par le modèle
+settings.video.negative_prompt = Invite négative
+settings.video.negative_prompt.desc = Mots ou phrases à éviter dans le résultat (séparés par des virgules)
+settings.video.prompt_model = Modèle d'amélioration de l'invite
+settings.video.prompt_model.desc = LLM utilisé pour affiner votre invite avant la génération de la vidéo (pas le modèle vidéo)
+settings.video.resolution = Résolution vidéo
+settings.video.resolution.desc = Résolution de sortie cible (ex. : 720p, 1080p); disponibilité dépend du modèle
+settings.video.seed = Graine
+settings.video.seed.desc = Graine aléatoire facultative pour des résultats reproductibles; laisser vide pour aléatoire
 settings.vision.capture.auto = Capture automatique
 settings.vision.capture.enabled = Caméra
 settings.vision.capture.height = Hauteur de capture par caméra (px)
@@ -1526,6 +1550,10 @@ updater.check.launch = Vérifier au démarrage
 update.released = build
 update.snap = Aller au Snap Store
 update.title = Vérification des mises à jour
+vid.status.downloading = Téléchargement de la vidéo... veuillez patienter...
+vid.status.generating = Génération de la vidéo à partir de
+vid.status.prompt.error = Erreur d'amélioration de l'invite survenue
+vid.status.prompt.wait = Préparation de l'invite... veuillez patienter...
 vision.capture.auto = Capture automatique
 vision.capture.auto.click = La capture automatique est activée !
 vision.capture.auto.label = Capture automatique activée

pygpt_net/data/locale/locale.it.ini

@@ -191,6 +191,7 @@ attachments_uploaded.tab = File caricati
 audio.cache.clear.confirm = Sei sicuro di voler eliminare tutti i file audio memorizzati nella cache?
 audio.cache.clear.success = OK. Tutti i file della cache audio sono stati cancellati.
 audio.control.btn = Controllo vocale
+audio.loop = Ciclo
 audio.magic_word.detected = Parola magica rilevata!
 audio.magic_word.invalid = Non è una parola magica :(
 audio.magic_word.please = Per favore, parola magica...
@@ -845,7 +846,7 @@ mode.agent_openai.tooltip = Agenti avanzati (OpenAI)
 mode.agent.tooltip = Agenti semplici (legacy)
 mode.assistant = Assistente
 mode.assistant.tooltip = Chattare utilizzando l'API degli Assistenti
-mode.audio = Chat con audio
+mode.audio = Realtime + audio
 mode.chat = Chat
 mode.chat.tooltip = Modalità chat (predefinita)
 mode.completion = Completamento
@@ -854,7 +855,7 @@ mode.computer = Uso del computer
 mode.computer.tooltip = Uso del computer (mouse, tastiera, navigazione)
 mode.expert = Esperti (cooperazione)
 mode.expert.tooltip = Chiamare esperti in background
-mode.img = Immagine
+mode.img = Immagine e video
 mode.img.tooltip = Generazione immagini con DALL-E
 mode.langchain = Langchain
 mode.langchain.tooltip = Chattare con modelli forniti da Langchain
@@ -1127,6 +1128,8 @@ settings.audio.input.stop_interval.desc = Intervallo in secondi per la trascrizi
 settings.audio.input.timeout = Timeout di registrazione
 settings.audio.input.timeout.continuous = Abilita timeout in modalità continua
 settings.audio.input.timeout.desc = Timeout (secondi) per l'arresto automatico della registrazione, 0 per disabilitare, predefinito: 120
+settings.audio.input.vad.prefix = Padding prefisso VAD (in ms)
+settings.audio.input.vad.silence = Silenzio finale VAD (in ms)
 settings.audio.output.backend = Backend per l'uscita audio
 settings.audio.output.backend.desc = Seleziona il backend per l'uscita audio.
 settings.audio.output.device = Dispositivo per l'uscita audio
@@ -1197,9 +1200,10 @@ settings.frequency_penalty = Penale di frequenza
 settings.func_call.native = Usa chiamate di funzione API native
 settings.func_call.native.desc = Se abilitato, l'applicazione utilizzerà le chiamate di funzione API native invece del formato interno pygpt e i prompt di comando di seguito non saranno utilizzati. Solo modalità chat e assistenti.
 settings.img_dialog_open = Apri la finestra di dialogo dell'immagine dopo la generazione (Modalità immagine)
-settings.img_prompt_model = DALL-E: modello di generazione del prompt
-settings.img_quality = DALL-E: qualità dell'immagine
-settings.img_resolution = DALL-E: dimensione dell'immagine
+settings.img_prompt_model = Modello di generazione del prompt
+settings.img_prompt_model.desc = LLM utilizzato per affinare il tuo prompt prima della generazione di immagini (non il modello di immagini)
+settings.img_quality = Qualità dell'immagine
+settings.img_resolution = Dimensione dell'immagine
 settings.layout.animation.disable = Disabilita animazioni
 settings.layout.animation.disable.desc = Disabilita le animazioni del layout, come i caricamenti animati, ecc.
 settings.layout.density = Densità del layout
@@ -1294,8 +1298,10 @@ settings.prompt.ctx.auto_summary.user = Contesto: auto-riassunto (messaggio dell
 settings.prompt.ctx.auto_summary.user.desc = Placeholder: {input}, {output}
 settings.prompt.expert = Esperto: Master prompt
 settings.prompt.expert.desc = Istruzione (prompt del sistema) per l'esperto Master su come gestire gli esperti subalterni. Le istruzioni per gli esperti subalterni sono date dalle loro preimpostazioni.
-settings.prompt.img = DALL-E: generazione immagine
+settings.prompt.img = Generazione immagine
 settings.prompt.img.desc = Prompt per generare comandi per DALL-E (se la modalità grezza è disabilitata). Solo modalità immagine.
+settings.prompt.video = Generazione video
+settings.prompt.video.desc = Prompt per generare prompt per il modello video (se la modalità raw è disattivata). Solo modalità Immagine/Video.
 settings.remote_tools.code_interpreter = Interprete del codice
 settings.remote_tools.code_interpreter.desc = Abilita l'attrezzo remoto `code_interpreter` in modalità Chat / tramite API delle Risposte di OpenAI.
 settings.remote_tools.file_search = File search
@@ -1339,7 +1345,9 @@ settings.section.ctx = Contesto
 settings.section.debug = Debug
 settings.section.files = File e allegati
 settings.section.general = Generale
-settings.section.images = Immagini
+settings.section.images = Immagini e video
+settings.section.images.image = Immagine
+settings.section.images.video = Video
 settings.section.layout = Layout
 settings.section.llama_index = Indici / LlamaIndex
 settings.section.llama-index.chat = Chat
@@ -1367,6 +1375,22 @@ settings.upload.data_dir.desc = Abilita per memorizzare tutto in una singola dir
 settings.upload.store = Conserva gli allegati nella cartella di upload del workdir
 settings.upload.store.desc = Abilita la conservazione di una copia locale degli allegati caricati per un uso futuro
 settings.use_context = Usa contesto (memoria)
+settings.video.aspect_ratio = Rapporto d'aspetto
+settings.video.aspect_ratio.desc = Rapporto d'aspetto del fotogramma (es. 16:9, 9:16, 1:1); la disponibilità dipende dal modello selezionato
+settings.video.duration = Durata video
+settings.video.duration.desc = Lunghezza della clip in secondi; i limiti possono variare in base al modello
+settings.video.fps = FPS
+settings.video.fps.desc = Fotogrammi al secondo (es. 24, 25, 30); può essere arrotondato o ignorato dal modello
+settings.video.generate_audio = Genera audio
+settings.video.generate_audio.desc = Include audio di sottofondo sintetizzato se supportato dal modello
+settings.video.negative_prompt = Prompt negativo
+settings.video.negative_prompt.desc = Parole o frasi da evitare nel risultato (separate da virgole)
+settings.video.prompt_model = Modello di miglioramento del prompt
+settings.video.prompt_model.desc = LLM utilizzato per affinare il tuo prompt prima della generazione video (non il modello video)
+settings.video.resolution = Risoluzione video
+settings.video.resolution.desc = Risoluzione di uscita target (es. 720p, 1080p); la disponibilità dipende dal modello
+settings.video.seed = Seme
+settings.video.seed.desc = Seme casuale opzionale per risultati riproducibili; lascia vuoto per casuale
 settings.vision.capture.auto = Cattura automatica
 settings.vision.capture.enabled = Fotocamera
 settings.vision.capture.height = Altezza cattura fotocamera (px)
@@ -1526,6 +1550,10 @@ updater.check.launch = Controlla all'avvio
 update.released = build
 update.snap = Vai allo Snap Store
 update.title = Controllo degli aggiornamenti
+vid.status.downloading = Download del video... si prega di attendere...
+vid.status.generating = Generazione video da
+vid.status.prompt.error = Si è verificato un errore nel prompt di miglioramento
+vid.status.prompt.wait = Preparazione del prompt... si prega di attendere...
 vision.capture.auto = Cattura automatica
 vision.capture.auto.click = La cattura automatica è abilitata!
 vision.capture.auto.label = Cattura automatica abilitata